ACM Contest Problems Archive

University of Valladolid (SPAIN)

363 Approximate Matches

Given a non-null string of text S , a non-null string of text P , and an integer k 0, nd all non-null
substrings of S that match or \approximately match" P with at most k mismatches. The substring S1
is considered to match the string P if they are identical (as you might expect), and approximately match
if the strings di er by no more than k mismatches, as de ned next.
If a minimum of m characters have to be removed from S1 to make it identical to P , then S1
approximately matches P with m mismatches. If a minimum of n characters have to be removed from P
to make it identical to S1 , then S1 approximately matches P with n mismatches. If m characters have
to be removed from S1 and n characters have to be removed from P to make S1 and P identical, then
S1 approximately matches P with m + n mismatches (here m + n should beminimized).
Consider the string S =\The sky is blue".
The substring \The sky" in S matches the string \The sky".
The substring \The sky" in S approximately matches the string \The skye" with one mismatch
(the extra \e" in the string).)
The substring \sky is" in S approximately matches the string \skis" with two mismatches (the
characters \y" and \ " in the substring).
The substring \sky is" in S approximately matches the string \tennis" with eight mismatches
(four characters are removed from \sky is" and four are removed from \tennis").
As another example, the string \scrapple" approximately matches the string \apples" with four
mismatches (by removing s, c, r from \scrapple" and s from \apples"). Although there are other
ways to remove letters from these two strings to make the results identical (for example removing all
but the s from both strings), four characters is the minimum number that can be removed.
Note that case is important in this problem; thus T does not match t, but T approximately matches
t with two mismatches.

Input

Each data set consists of three lines that contain k, S , and P , respectively. k is an integer; S is a string
of length between 1 and 50 inclusive, and P is a string of length between 1 and 20 inclusive. Your
program must stop processing data when it encounters a negative value for k

Output

For each match with i mismatches (0 i k), your program must generate a line of one of the following
forms (whichever is appropriate):
Q matches P
Q matches P with 1 mismatch
Q matches P with i mismatches
where Q is a substring of S that matches P or that approximately matches P with i mismatches. If
more than one substring in S matches or approximately matches P , they must be printed in the order
they appear in S (from left to right). Substrings that start at the same position must be listed in the

ACM Contest Problems Archive

University of Valladolid (SPAIN)

order of their size (shortest to longest). No value for Q may be listed more than once. If there is no
match, no output should be generated. Leave a blank line after the output for each data set (even if
there is no output for the data set).

Sample Input
1
He did not care about the carpet in the car.
car
2
ABC
BC
-1

Sample Output
car matches car with 1 mismatch
ca matches car with 1 mismatch
car matches car
care matches car with 1 mismatch
ar matches car with 1 mismatch
carp matches car with 1 mismatch
car. matches car with 1 mismatch
AB matches BC with 2 mismatches
ABC matches BC with 1 mismatch
B matches BC with 1 mismatch
BC matches BC
C matches BC with 1 mismatch